Method, device, equipment and medium for traffic increment prediction of content distribution network

Abstract

This invention provides a method, apparatus, device, and medium for predicting service increments in a content delivery network. The method involves acquiring bandwidth data from multiple historical periods for a target service; selecting the most recent historical period's bandwidth data as reference bandwidth data from the multiple historical periods; determining the similarity between the reference bandwidth data and bandwidth data from other historical periods; determining a preset number of target historical periods' bandwidth data based on the similarity; performing a weighted average on the target historical period's bandwidth data to determine the estimated bandwidth data for the next period; and determining the target increment information for the next period based on the estimated bandwidth data for the next period. By utilizing bandwidth data from multiple historical periods to filter out the target historical period's bandwidth data and determining the target increment information based on the target historical period's bandwidth data, the efficiency and accuracy of predicting the target increment information for the next period are improved.

Description

Technical Field

[0001] This invention relates to the field of network communication technology, and in particular to a method, apparatus, device, and medium for predicting service increments in a content delivery network. Background Technology

[0002] With the booming development of the digital economy and the widespread adoption of the internet, the Content Delivery Network (CDN) market has experienced unprecedented rapid growth. The number of customers served by CDN providers continues to climb, and their business scale is constantly expanding. Against this backdrop, CDN providers need to allocate differentiated bandwidth resources to meet the diverse business needs of different customers to ensure a high-quality experience for CDN acceleration services. For large customers, the end of each month often brings significant increases in business demand. To effectively address the challenges of limited total resources and internal cost control while ensuring stable business operations, accurately assessing the resource needs for the following month becomes particularly crucial.

[0003] Traditional methods typically rely on manual collection of incremental customer service information in advance, followed by resource allocation based on this information, and then manual inspection of bandwidth records after the incremental increase to compare with the actual increase in service volume. However, this method is labor-intensive and time-consuming, and its efficiency and accuracy are relatively low. Summary of the Invention

[0004] In view of the above problems, embodiments of the present invention are proposed to provide a method, apparatus, device and medium for predicting service increments in a content delivery network that overcomes or at least partially solves the above problems.

[0005] To address the aforementioned problems, this invention discloses a method for predicting service increments in a content delivery network, the method comprising:

[0006] Acquire bandwidth data for multiple historical periods of the target service; wherein the bandwidth data for each historical period includes bandwidth data collected at multiple sampling times according to a preset sampling interval;

[0007] Based on the bandwidth data from the multiple historical periods, the bandwidth data from the most recent historical period is selected as the reference bandwidth data;

[0008] Determine the similarity between the reference bandwidth data and bandwidth data from other historical periods besides the reference bandwidth data;

[0009] Based on the similarity, a preset number of bandwidth data for a target historical period are determined;

[0010] The bandwidth data for the target historical period is weighted and averaged to determine the estimated bandwidth data for the next period;

[0011] Based on the estimated bandwidth data for the next cycle, the target increment information for the next cycle is determined.

[0012] Optionally, determining the similarity between the reference bandwidth data and bandwidth data from other historical periods besides the reference bandwidth data includes:

[0013] Based on the dynamic time warping function, a first distance is calculated between the reference bandwidth data and the bandwidth data of other historical periods; the first distance is inversely correlated with the similarity.

[0014] Based on the first distance, the similarity between the reference bandwidth data and bandwidth data from other historical periods is determined.

[0015] Optionally, calculating the first distance between the reference bandwidth data and the bandwidth data of other historical periods based on the dynamic time warping function includes:

[0016] Determine the sub-bandwidth data of the reference bandwidth data;

[0017] Based on the dynamic time warping function, calculate the second distance between the sub-bandwidth data of the reference bandwidth data and the bandwidth data of other historical periods;

[0018] If the second distance is less than or equal to the first preset threshold, then the reference bandwidth data is calculated, and the first distance between the reference bandwidth data and the bandwidth data of other historical periods corresponding to the second distance is calculated.

[0019] Optionally, determining a preset number of bandwidth data for a target historical period based on the similarity includes:

[0020] Determine the bandwidth data of the candidate historical period corresponding to the first distance that is less than or equal to the second preset threshold;

[0021] The bandwidth data of the candidate historical periods are sorted in ascending order of the first distance;

[0022] A preset number of candidate historical periods' bandwidth data are selected in the order described above, and these are used as the bandwidth data for the target historical period.

[0023] Optionally, determining the target increment information for the next period based on the estimated bandwidth data for the next period includes:

[0024] Based on the estimated bandwidth data for the next cycle, determine the first estimated increment value;

[0025] Obtain the second estimated increment value;

[0026] The target increment information for the next period is determined based on the first estimated increment value and the second estimated increment value.

[0027] Optionally, determining the target increment information for the next period based on the first estimated increment value and the second estimated increment value includes:

[0028] Obtain the estimated accuracy value and the business risk value;

[0029] The target increment information for the next period is determined based on the estimated accuracy value, the business risk value, the first estimated increment value, and the second estimated increment value.

[0030] Optionally, before selecting the bandwidth data from the most recent historical period as the reference bandwidth data based on the bandwidth data from the multiple historical periods, the method further includes:

[0031] For bandwidth data of each historical period, if the bandwidth at the sampling time is less than the third preset threshold, the bandwidth at the previous sampling time is greater than or equal to the third preset threshold, and the bandwidth at the next sampling time is greater than or equal to the third preset threshold, then the bandwidth at the sampling time is determined to be abnormal. The bandwidth at the sampling time is determined by interpolation based on the bandwidth at the previous sampling time and the bandwidth at the next sampling time.

[0032] If the bandwidth at the sampling time is greater than the fourth preset threshold, and the ratio of the difference between the bandwidth at the sampling time and the bandwidth at the previous sampling time to the bandwidth at the previous sampling time is greater than the fourth preset threshold, then the bandwidth at the sampling time is determined to be a sudden drop value, and the bandwidth at the sampling time is smoothed.

[0033] Accordingly, embodiments of the present invention disclose a service increment prediction device for a content delivery network, the device comprising:

[0034] A bandwidth data acquisition module is used to acquire bandwidth data for multiple historical periods of a target service; wherein the bandwidth data for each historical period includes bandwidth data collected at multiple sampling times according to a preset sampling interval.

[0035] The reference bandwidth data determination module is used to select the bandwidth data of the most recent historical period as the reference bandwidth data based on the bandwidth data of the multiple historical periods.

[0036] A similarity determination module is used to determine the similarity between the reference bandwidth data and bandwidth data from other historical periods other than the reference bandwidth data;

[0037] The target historical period bandwidth data determination module is used to determine a preset number of target historical period bandwidth data based on the similarity.

[0038] The estimated bandwidth data determination module is used to perform a weighted average of the bandwidth data of the target historical period to determine the estimated bandwidth data for the next period.

[0039] The target increment information determination module is used to determine the target increment information for the next period based on the estimated bandwidth data for the next period.

[0040] Optionally, the similarity determination module includes:

[0041] The first distance determination submodule is used to calculate a first distance between the reference bandwidth data and the bandwidth data of other historical periods based on a dynamic time warping function; the first distance is inversely correlated with the similarity.

[0042] The similarity determination submodule is used to determine the similarity between the reference bandwidth data and bandwidth data from other historical periods other than the reference bandwidth data, based on the first distance.

[0043] Optionally, the first distance determination submodule includes:

[0044] A sub-bandwidth data determination unit is used to determine the sub-bandwidth data of the reference bandwidth data;

[0045] The second distance determination unit is used to calculate, according to the dynamic time warping function, the second distance between the sub-bandwidth data of the reference bandwidth data and the bandwidth data of other historical periods;

[0046] The first distance determination unit is used to calculate the first distance between the reference bandwidth data and the bandwidth data of other historical periods corresponding to the second distance if the second distance is less than or equal to the first preset threshold.

[0047] Optionally, the bandwidth data determination module for the target historical period includes:

[0048] The bandwidth data determination submodule for candidate historical periods is used to determine the bandwidth data of the candidate historical period corresponding to a first distance that is less than or equal to a second preset threshold.

[0049] The sorting submodule is used to sort the bandwidth data of the candidate historical periods in ascending order of the first distance;

[0050] The target historical period bandwidth data determination submodule is used to select a preset number of candidate historical periods' bandwidth data in the order described above, as the bandwidth data for the target historical period.

[0051] Optionally, the target incremental information determination module includes:

[0052] The first estimated increment value determination submodule is used to determine the first estimated increment value based on the estimated bandwidth data of the next cycle.

[0053] The second estimated increment value acquisition submodule is used to acquire the second estimated increment value;

[0054] The target increment information determination submodule is used to determine the target increment information for the next period based on the first estimated increment value and the second estimated increment value.

[0055] Optionally, the target incremental information determination submodule includes:

[0056] The data acquisition unit is used to acquire the estimated accuracy value and the business risk value;

[0057] The target increment information determination unit is used to determine the target increment information for the next period based on the estimated accuracy value, the business risk value, the first estimated increment value, and the second estimated increment value.

[0058] Optionally, the device further includes:

[0059] An interpolation processing module is used to determine if the bandwidth at the sampling time is abnormal for bandwidth data of each historical period. If the bandwidth at the sampling time is less than a third preset threshold, the bandwidth at the previous sampling time is greater than or equal to the third preset threshold, and the bandwidth at the next sampling time is greater than or equal to the third preset threshold, then the bandwidth at the sampling time is determined to be abnormal. The module then interpolates based on the bandwidth at the previous sampling time and the bandwidth at the next sampling time to determine the bandwidth at the sampling time.

[0060] A smoothing module is used to determine that the bandwidth at the sampling time is a sudden drop value if the bandwidth at the sampling time is greater than a fourth preset threshold, and the ratio of the difference between the bandwidth at the sampling time and the bandwidth at the previous sampling time to the bandwidth at the previous sampling time is greater than the fourth preset threshold, and then smooths the bandwidth at the sampling time.

[0061] Accordingly, this invention discloses an electronic device, including: a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, it implements the various steps of the above-described embodiment of the service increment prediction method for a content delivery network.

[0062] Accordingly, this invention discloses a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the various steps of the above-described embodiment of the service increment prediction method for a content delivery network.

[0063] The embodiments of the present invention have the following advantages:

[0064] The service increment prediction method for a content delivery network according to embodiments of the present invention acquires bandwidth data for multiple historical periods of a target service. Each historical period's bandwidth data includes bandwidth data collected at multiple sampling times according to a preset sampling interval. From the bandwidth data of multiple historical periods, the most recent historical period's bandwidth data is selected as reference bandwidth data. The similarity between the reference bandwidth data and bandwidth data from other historical periods is determined. Based on the similarity, a preset number of target historical period's bandwidth data are determined. A weighted average is applied to the bandwidth data of the target historical periods to determine the estimated bandwidth data for the next period. Based on the estimated bandwidth data for the next period, the target increment information for the next period is determined. By utilizing bandwidth data from multiple historical periods to filter out the target historical period's bandwidth data and determining the target increment information based on the target historical period's bandwidth data, the efficiency and accuracy of predicting the target increment information for the next period are improved. Attached Figure Description

[0065] Figure 1 This is a flowchart illustrating the steps of a method for predicting incremental service usage in a content delivery network according to an embodiment of the present invention.

[0066] Figure 2 This is a schematic diagram of another method for predicting the incremental business of a content delivery network according to the present invention;

[0067] Figure 3 This is a structural block diagram of a device for predicting the incremental service volume of a content delivery network according to an embodiment of the present invention. Detailed Implementation

[0068] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0069] With the booming development of the digital economy and the widespread adoption of the internet, the content delivery network (CDN) market has experienced unprecedented rapid growth. The number of customers served by CDN providers continues to climb, and their business scale is constantly expanding. Against this backdrop, CDN providers need to allocate differentiated bandwidth resources to meet the diverse business needs of different customers to ensure a high-quality experience for CDN acceleration services. For large customers, the end of each month often brings significant increases in business demand. To effectively address the challenges of limited total resources and internal cost control while ensuring stable business operations, accurately assessing the resource needs for the following month becomes particularly crucial.

[0070] Traditional methods typically rely on manual collection of incremental customer service information in advance, followed by resource allocation based on this information, and then manual inspection of bandwidth records after the incremental increase to compare with the actual increase in service volume. However, this method is labor-intensive and time-consuming, and its efficiency and accuracy are relatively low.

[0071] One of the core concepts of this invention is that by using bandwidth data from multiple historical periods to filter out bandwidth data for a target historical period, and determining target incremental information based on the bandwidth data of the target historical period, the efficiency and accuracy of predicting target incremental information for the next period are improved.

[0072] Reference Figure 1 The diagram illustrates a flowchart of a method for predicting service increments in a content delivery network according to an embodiment of the present invention, which may specifically include the following steps:

[0073] Step 101: Obtain bandwidth data for multiple historical periods of the target service; wherein, the bandwidth data for each historical period includes bandwidth data collected at multiple sampling times according to a preset sampling interval.

[0074] Specifically, if a customer's business is incrementally operating on the content delivery network, and there is at least one historical period, historical bandwidth data for multiple historical periods of the target customer's business can be obtained based on the customer's name or domain name. For example, bandwidth data Y for K historical periods can be obtained. his =(Y1,Y2,…,Y) k In the i-th historical period, there are N sampling times, and the bandwidth data of one historical period is Y. i =(y i1 ,y i2 ,…,y iN ), where 1≤i≤K, and K can be set according to actual needs.

[0075] A historical period duration T can be 1 month, and the sampling interval tp can be 1 minute, 5 minutes, 1 hour, etc. For example, if the historical period duration T is 1 month (30 days) and the sampling interval tp is 5 minutes, then theoretically, 8640 sampling moments of historical bandwidth data can be obtained in one historical period. Of course, the historical period duration and sampling interval can be set according to actual needs in practical applications, and the embodiments of the present invention do not limit this.

[0076] Step 102: Based on the bandwidth data of the multiple historical periods, select the bandwidth data of the most recent historical period as the reference bandwidth data.

[0077] Specifically, based on bandwidth data from multiple historical periods, the bandwidth data from the historical period closest to the current time is selected as the reference bandwidth data, which is the matching sample Y0.

[0078] Step 103: Determine the similarity between the reference bandwidth data and bandwidth data from other historical periods besides the reference bandwidth data.

[0079] Specifically, after selecting reference bandwidth data Y0, the similarity between reference bandwidth data Y0 and bandwidth data from other historical periods is calculated.

[0080] Step 104: Based on the similarity, determine a preset number of bandwidth data for the target historical period.

[0081] Specifically, after determining the similarity between the reference bandwidth data Y0 and the bandwidth data of other historical periods, a preset number of target historical periods' bandwidth data are determined from the bandwidth data of other historical periods based on the similarity. For example, w target historical periods' bandwidth data are determined. Of course, the preset number can be set according to actual needs in practical applications, and this embodiment of the invention does not limit this.

[0082] In actual business operations, there may be situations where there is no incremental data in a certain historical period, or only incremental data in certain periods, or only a small amount of grayscale data in a certain period. Therefore, it is necessary to first filter out the bandwidth data of the target historical period from the bandwidth data of multiple historical periods, and then make subsequent incremental estimates to improve the accuracy of the estimates.

[0083] Step 105: Perform a weighted average of the bandwidth data for the target historical period to determine the estimated bandwidth data for the next period.

[0084] Specifically, the bandwidth data for each sampling moment in the next period is obtained by weighted averaging of the bandwidth data from the w target historical periods, which can be calculated using the following formula:

[0085]

[0086] Where Y′ represents the estimated bandwidth data, and D i As the weight, Y d,i For the bandwidth data of the target historical period, the weight corresponding to each target historical period is determined based on the similarity between the reference bandwidth data Y0 and the bandwidth data of the target historical period. The greater the similarity, the greater the weight of the bandwidth data of the target historical period.

[0087] Step 106: Determine the target increment information for the next period based on the estimated bandwidth data for the next period.

[0088] Specifically, after determining the estimated bandwidth data Y′ for the next period, the estimated peak value, 95% value, and other incremental information for the next period can be determined based on the estimated bandwidth data Y′. Based on the estimated peak value, 95% value, and other incremental information for the next period, the target incremental information for the next period can be determined.

[0089] In content delivery network (CDN) business analytics, peak traffic and the 95th percentile are two important performance metrics used to measure network traffic and service quality. Peak traffic refers to the highest point of network traffic or service requests within a given timeframe, reflecting the system's load during peak hours. The 95th percentile refers to the 95th percentile of network traffic or service requests within a given timeframe. It indicates that for 95% of the time, traffic or requests are below this value. The 95th percentile is typically used to evaluate the system's average performance, especially when handling fluctuating traffic volumes.

[0090] In this embodiment of the invention, step 103, determining the similarity between the reference bandwidth data and bandwidth data from other historical periods, may specifically include the following sub-steps:

[0091] Sub-step S11: Calculate the first distance between the reference bandwidth data and the bandwidth data of other historical periods according to the dynamic time warping function; the first distance is inversely correlated with the similarity.

[0092] Specifically, the historical period is one month long. Since the number of days in each month is not necessarily the same, the number of samples in each historical period may be different. In order to solve the problem of unequal lengths of two historical period sequences, the dynamic time warping function can be selected as the similarity measurement function D between the historical period bandwidth sequences. Dynamic time warping (DTW) is an algorithm used to measure the similarity between two time series. It is particularly suitable for time series data that have offsets, scaling or distortions on the time axis. DTW finds the best matching path between two sequences by performing non-linear alignment on the time axis, thereby calculating the distance or similarity between them.

[0093] Based on the dynamic time warping function, the first distance between the reference bandwidth data Y0 and the bandwidth data of other historical periods is calculated. The first distance is D(Y0,Y), where D is the dynamic time warping function and Y is the bandwidth data sequence of other historical periods. The smaller the first distance, the greater the similarity.

[0094] Sub-step S12: Based on the first distance, determine the similarity between the reference bandwidth data and bandwidth data from other historical periods besides the reference bandwidth data.

[0095] Specifically, after determining the first distance between the reference bandwidth data Y0 and the bandwidth data of other historical periods, since the smaller the first distance, the greater the similarity, the similarity between the reference bandwidth data Y0 and the bandwidth data of other historical periods is determined based on the first distance.

[0096] In this embodiment of the invention, sub-step S11 calculates a first distance between the reference bandwidth data and the bandwidth data of other historical periods based on a dynamic time warping function; the first distance is inversely correlated with the similarity, and may specifically include the following sub-steps:

[0097] Sub-step S111: Determine the sub-bandwidth data of the reference bandwidth data.

[0098] Specifically, the reference bandwidth data Y0 = (y 01 ,y 02 ,…,y 0N ), Select a continuous temporal subset Y0[1:jgap] = {y from the reference bandwidth data Y0. 01 ,y 01 ,…,y 0jgap}, which is the sub-bandwidth data sequence of Y0, denoted as P, that is, P is a time series of length jgap.

[0099] Sub-step S112: Calculate the second distance between the sub-bandwidth data of the reference bandwidth data and the bandwidth data of other historical periods according to the dynamic time warping function.

[0100] Specifically, based on the dynamic time warping function, the second distance between the sub-bandwidth data of the reference bandwidth data and the bandwidth data of other historical periods is calculated, where the second distance is D(P,Q).

[0101] Sub-step S113: If the second distance is less than or equal to the first preset threshold, then calculate the reference bandwidth data and the first distance between the bandwidth data of other historical periods corresponding to the second distance.

[0102] Specifically, the first preset threshold ε′ is a constant greater than 0, which can be set according to actual needs. If D=(P,Q)≤ε′, then the corresponding sequence Q in the historical periodic bandwidth data at this time is recorded and the subsequent distance calculation is continued; if D=(P,Q) is greater than ε′, then the search for the remaining sequences in the historical periodic bandwidth data is stopped.

[0103] To reduce the number of sequences in historical periodic data that need to be matched with the reference bandwidth data Y0, thereby reducing the computational load of dynamic time warping distance, fast matching is achieved by extracting subsequences.

[0104] In this embodiment of the invention, step 104 determines a preset number of bandwidth data for a target historical period based on the similarity, which may specifically include the following sub-steps:

[0105] Sub-step S21: Determine the bandwidth data of the candidate historical period corresponding to the first distance which is less than or equal to the second preset threshold.

[0106] Specifically, the second preset threshold is ε, and the bandwidth data of candidate historical periods corresponding to the first distance D(Y0,Y) less than or equal to ε are determined. The set of bandwidth data sequences of candidate historical periods is as follows:

[0107] Sub-step S22: Sort the bandwidth data of the candidate historical periods in ascending order of the first distance.

[0108] Specifically, the first distances that satisfy the condition of being less than or equal to ε are sorted in ascending order, that is, in descending order of similarity.

[0109] Sub-step S23: Select a preset number of bandwidth data from candidate historical periods in the order described above, and use them as the bandwidth data for the target historical period.

[0110] Specifically, following the order of the first distance from smallest to largest, the bandwidth data of the candidate historical periods corresponding to the first distances are taken as the bandwidth data of the target historical period, denoted as Y. d,i (1≤i≤w).

[0111] In this embodiment of the invention, step 106, based on the estimated bandwidth data for the next cycle, determines the target increment information for the next cycle, which may specifically include the following sub-steps:

[0112] Sub-step S31: Determine the first estimated increment value based on the estimated bandwidth data for the next cycle.

[0113] Specifically, based on the estimated bandwidth data Y′ for the next period, the estimated peak value, 95% value, and other incremental information for the next period are determined. This estimated peak value, 95% value, and other incremental information for the next period are used as the first estimated value in the final output, denoted as Y′.

[0114] Sub-step S32: Obtain the second estimated increment value.

[0115] Specifically, in actual business interactions, if the business side or the client provides incremental information for the next period in advance (such as peak values, 95% values, etc.), this can be used as a second estimated value, denoted as [value 1], [value 2], [value 3], [value 4], [value 5], [value 6], [value 7], [value 8], [value 9], [value 10], [value 11], [value 12], [value 13], [value 14], [value 15], [value 16], [value 17], [value 18], [value 19 ...

[0116] Sub-step S33: Determine the target increment information for the next period based on the first estimated increment value and the second estimated increment value.

[0117] Specifically, the first estimate is revised using the second estimate to output the final estimated increment information Y for the next period. p .

[0118] In this embodiment of the invention, sub-step S33 determines the target increment information for the next period based on the first estimated increment value and the second estimated increment value, which may specifically include the following sub-steps:

[0119] Sub-step S331: Obtain the estimated accuracy value and business risk value.

[0120] Specifically, the prediction accuracy value is the accuracy of the predicted increment value. A time frame can be set to statistically analyze the current business increment. Let's assume the actual business increment is Y. R The accuracy 'a' of the estimated value can be evaluated using the following formula:

[0121]

[0122] After each increment, the accuracy of the estimated increment for that business is updated. Assuming there have been n increments, then 'a' is updated as follows:

[0123]

[0124] Where λ is the weight of the first estimated value after the increment, and λ1 < λ2 < ... < λ n The weight value can be preset to any number within (0,1], and the closer to the current time, the greater the corresponding weight.

[0125] The business risk value 'b' can be calculated based on the estimated accuracy value, combined with multiple dimensions such as business configuration complexity (including the number of domains involved in the incremental changes, the number of development items, etc.), the cooperation of business contacts, and the number of historical complaints. Each evaluation dimension has a preset set of proportional coefficients. The corresponding proportional coefficient is determined from this preset set, and finally, the estimated accuracy value and the proportional coefficient for each dimension are added together to obtain the final business risk value. Taking the evaluation dimension of the number of domains as an example, assuming the proportional coefficient is β1 for the number of domains in the range [0,10], β2 for the number of domains in the range [10,50], and so on, the proportional coefficients corresponding to different numbers of domains can be obtained.

[0126] Similarly, the risk coefficient for that business is updated after each increment.

[0127] Sub-step S332: Determine the target increment information for the next period based on the estimated accuracy value, the business risk value, the first estimated increment value, and the second estimated increment value.

[0128] Specifically, the target increment information for the next period can be determined using the following formula:

[0129]

[0130] Where a is the estimated accuracy value and b is the business risk value corresponding to this business.

[0131] The business risk value is initialized to 1. The values ​​of 'a' and 'b' need to be updated after each subsequent business increment. It should be noted that if this is the first increment, the first estimated increment value cannot be obtained from historical period data. If the business or client side can provide second estimated value information, it will be directly confirmed according to the formula for determining the target increment information for the next period. In this case, 'a' = 1, and the target increment information for the next period is...

[0132] When acquiring bandwidth data for multiple historical periods of a target service, information such as the service's historical risk coefficient, service type, service coefficient, and whether special resource coverage is used can also be obtained. If the service has special coverage resource requirements, this method will not be used for incremental prediction. Instead, resources will be evaluated separately. For example, some services have low settlement prices and customers may be highly sensitive to business needs. Therefore, it is necessary to communicate with customers to adjust resources. Under the premise that our cost-benefit ratio is appropriate, special resource coverage will be adopted in some regions.

[0133] After this incremental allocation is completed, the risk coefficients of different services will be sorted in ascending order, and different resource priority tiers will be assigned based on the risk coefficients of each service. The higher the risk coefficient, the lower the resource preparation priority. For example, services with the highest risk coefficients can be associated with high-priority tier tags, the lowest 30% can be associated with low-priority tier tags, and the remaining services can be associated with medium-priority tier tags. The total platform bandwidth resources are limited and may not be able to meet the incremental needs of all customers; therefore, resource assessment needs to allocate resources to different customers according to certain priorities.

[0134] In this embodiment of the invention, before selecting the bandwidth data of the most recent historical period as the reference bandwidth data based on the bandwidth data of the plurality of historical periods, the method further includes:

[0135] For bandwidth data of each historical period, if the bandwidth at the sampling time is less than a third preset threshold, the bandwidth at the previous sampling time is greater than or equal to the third preset threshold, and the bandwidth at the next sampling time is greater than or equal to the third preset threshold, then the bandwidth at the sampling time is determined to be abnormal. The bandwidth at the sampling time is determined by interpolation based on the bandwidth at the previous sampling time and the bandwidth at the next sampling time.

[0136] Specifically, after obtaining bandwidth data for multiple historical periods of the target service, it is necessary to preprocess the bandwidth data for multiple historical periods, mainly including two parts: single point missing values ​​and burst (decrease) values.

[0137] A bandwidth threshold δ2 can be set if the bandwidth y(t) corresponding to sampling time t satisfies the following relationship:

[0138]

[0139] If the data at that sampling time is considered to be abnormally missing, it may be due to platform failure or other reasons. In this case, the bandwidth at that sampling time can be obtained by interpolating the data from the two sampling times before and after. The sampling interval is generally on the order of minutes. The bandwidth data of adjacent sampling times is usually relatively smooth, and common linear interpolation methods can be used. Of course, the interpolation method can be set according to actual needs in practical applications, and this embodiment of the invention does not limit it.

[0140] If the bandwidth at the sampling time is greater than the fourth preset threshold, and the ratio of the difference between the bandwidth at the sampling time and the bandwidth at the previous sampling time to the bandwidth at the previous sampling time is greater than the fourth preset threshold, then the bandwidth at the sampling time is determined to be a sudden drop value, and the bandwidth at the sampling time is smoothed.

[0141] Specifically, handling sudden (decrease) values ​​assumes that the business has experienced at least one incremental change. This can be divided into two scenarios: expected sudden changes and unexpected sudden changes.

[0142] Expected bursts (decreases): After the increment, business characteristics can be obtained through manual analysis. For example, a certain business may experience bandwidth bursts during holidays (such as New Year's Day, Spring Festival Gala, etc.). At this time, the bandwidth burst phenomenon during the corresponding period is expected. Therefore, the sampled bandwidth data of this period in the historical period can be manually marked and retained.

[0143] Unexpected sudden drops (decreases): For example, a service experiences a sudden drop or even a complete loss in bandwidth over a period of time, but after contacting the customer, it is confirmed that this is caused by a bug in the customer's system. In this case, the bandwidth drop during that period is unexpected, and therefore the historical data for that period needs to be manually marked for subsequent smoothing. If manual marking is not possible beforehand, the percentage change in bandwidth can be used to determine whether it is a sudden drop (decrease) according to the following rules:

[0144]

[0145] Wherein, δ3 is the set bandwidth mutation rate threshold, and δ4 is the set bandwidth threshold. It should be noted that δ3 is related to the service type; the bandwidth mutation rate threshold may differ for different services. When the bandwidth at a given sampling moment is greater than the threshold δ4 and the bandwidth change rate is greater than the set threshold δ3, the bandwidth data at that moment is considered a burst (decrease), and the data needs to be smoothed.

[0146] Reference Figure 2 The diagram illustrates another method for predicting incremental business in a content delivery network according to the present invention. The method includes a data acquisition and recording module, a data preprocessing module, a data prediction module, a data analysis module, and a data display module.

[0147] The data acquisition and recording module records historical periodic data and business information, providing a data foundation for the data preprocessing module. It should be noted that in the context of incremental business in content delivery networks, the period in this paper is generally one month. The data preprocessing module preprocesses the collected historical periodic bandwidth data according to certain business characteristic rules, including handling single-point missing values ​​and burst (decrease) values ​​in the collected bandwidth data. The data prediction module, considering the actual business scenario, filters and quickly matches the target historical period from multiple preprocessed historical periods, then constructs a prediction model to obtain the first estimated value of the incremental information for the next period. Furthermore, it combines incremental information provided by the business or customer side as a second estimated value, and finally estimates the predicted incremental information for the next period by combining the first and second estimated values. The data analysis module compares the incremental estimated value with the actual incremental situation to obtain the accuracy of the estimated value, analyzing whether the business incremental meets expectations. Based on the accuracy of the estimated value and other dimensions such as business configuration complexity, a business risk coefficient is calculated. After the incremental period ends, the risk coefficients of each business that increased volume in this period are sorted and categorized to associate different resource priority tags. The data visualization module integrates the analysis results from the data analysis module (including information such as business type, business coefficient, risk coefficient, and resource priority) and outputs relevant reports, which are then sent to relevant operations personnel for subsequent business opportunity assessment and resource preparation. Finally, the analysis results are uploaded and updated again to the business management platform, generating an analysis report based on the above modules. The report includes customer account and email, business name, corresponding operations personnel and contact information, front-end personnel name and contact information, domain name, edge and parent node coverage scheme names, business configuration complexity (including the number of domain names and development items involved in the incremental change), cooperation status of business contacts, historical complaints, accuracy of the estimated value, risk coefficient, and resource priority tags. After the incremental change is completed, the analysis report is sent to the corresponding operations personnel and resource colleagues to better prepare resource redundancy and assess the accuracy of business opportunities in the next cycle.

[0148] The method for predicting service increments in a content delivery network according to embodiments of the present invention has the following advantages:

[0149] It is more practical. Considering actual business scenarios, the collected historical data is preprocessed according to business characteristic rules; based on the similarity of the business cycle waveform of the content delivery network, the target historical cycle is selected by fully utilizing historical cycle data; in addition to obtaining the estimated value based on the historical cycle, it is also corrected by combining incremental information provided by the business or customer side, which improves the accuracy of the estimated incremental information; the business risk coefficient is calculated through multiple dimensions, sorted, and associated with different priority tags.

[0150] It is more adaptive. It makes full use of data from multiple business increments to continuously update incremental estimates, the accuracy of estimates, and business risk coefficients.

[0151] It requires less computation. It achieves rapid matching of target historical periods by extracting subsequences of periodic bandwidth data.

[0152] Incremental automated analysis solves the problem of low efficiency in manually analyzing incremental data, thus providing a reference for preparing resource redundancy for different customer businesses, which helps to improve the rationality of resource allocation and reduce enterprise costs.

[0153] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of the present invention are not limited to the described order of actions, because according to the embodiments of the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions involved are not necessarily essential to the embodiments of the present invention.

[0154] Reference Figure 3 The diagram illustrates a structural block diagram of an embodiment of a service increment prediction device for a content delivery network according to the present invention, which may specifically include the following modules:

[0155] The bandwidth data acquisition module 201 is used to acquire bandwidth data for multiple historical periods of the target service; wherein the bandwidth data for each historical period includes bandwidth data at multiple sampling times collected according to a preset sampling interval.

[0156] The reference bandwidth data determination module 202 is used to select the bandwidth data of the most recent historical period as the reference bandwidth data based on the bandwidth data of the multiple historical periods.

[0157] The similarity determination module 203 is used to determine the similarity between the reference bandwidth data and bandwidth data from other historical periods besides the reference bandwidth data.

[0158] The target historical period bandwidth data determination module 204 is used to determine a preset number of target historical period bandwidth data based on the similarity.

[0159] The estimated bandwidth data determination module 205 is used to perform a weighted average of the bandwidth data of the target historical period to determine the estimated bandwidth data for the next period.

[0160] The target increment information determination module 206 is used to determine the target increment information for the next period based on the estimated bandwidth data for the next period.

[0161] The service increment prediction device for a content delivery network according to this invention acquires bandwidth data for multiple historical periods of a target service. Each historical period's bandwidth data includes bandwidth data collected at multiple sampling times according to a preset sampling interval. From the bandwidth data of multiple historical periods, the device selects the bandwidth data of the most recent historical period as reference bandwidth data. It then determines the similarity between the reference bandwidth data and the bandwidth data of other historical periods. Based on the similarity, it determines a preset number of bandwidth data for a target historical period. Finally, it performs a weighted average of the bandwidth data for the target historical period to determine the estimated bandwidth data for the next period. Based on the estimated bandwidth data for the next period, it determines the target increment information for the next period. By using bandwidth data from multiple historical periods to filter out the bandwidth data for the target historical period and determining the target increment information based on the bandwidth data of the target historical period, the efficiency and accuracy of predicting the target increment information for the next period are improved.

[0162] In this embodiment of the invention, the similarity determination module includes:

[0163] The first distance determination submodule is used to calculate a first distance between the reference bandwidth data and the bandwidth data of other historical periods based on a dynamic time warping function; the first distance is inversely correlated with the similarity.

[0164] The similarity determination submodule is used to determine the similarity between the reference bandwidth data and bandwidth data from other historical periods other than the reference bandwidth data, based on the first distance.

[0165] In this embodiment of the invention, the first distance determination submodule includes:

[0166] A sub-bandwidth data determination unit is used to determine the sub-bandwidth data of the reference bandwidth data;

[0167] The second distance determination unit is used to calculate, according to the dynamic time warping function, the second distance between the sub-bandwidth data of the reference bandwidth data and the bandwidth data of other historical periods;

[0168] The first distance determination unit is used to calculate the first distance between the reference bandwidth data and the bandwidth data of other historical periods corresponding to the second distance if the second distance is less than or equal to the first preset threshold.

[0169] In this embodiment of the invention, the bandwidth data determination module for the target historical period includes:

[0170] The bandwidth data determination submodule for candidate historical periods is used to determine the bandwidth data of the candidate historical period corresponding to a first distance that is less than or equal to a second preset threshold.

[0171] The sorting submodule is used to sort the bandwidth data of the candidate historical periods in ascending order of the first distance;

[0172] The target historical period bandwidth data determination submodule is used to select a preset number of candidate historical periods' bandwidth data in the order described above, as the bandwidth data for the target historical period.

[0173] In this embodiment of the invention, the target incremental information determination module includes:

[0174] The first estimated increment value determination submodule is used to determine the first estimated increment value based on the estimated bandwidth data of the next cycle.

[0175] The second estimated increment value acquisition submodule is used to acquire the second estimated increment value;

[0176] The target increment information determination submodule is used to determine the target increment information for the next period based on the first estimated increment value and the second estimated increment value.

[0177] In this embodiment of the invention, the target incremental information determination submodule includes:

[0178] The data acquisition unit is used to acquire the estimated accuracy value and the business risk value;

[0179] The target increment information determination unit is used to determine the target increment information for the next period based on the estimated accuracy value, the business risk value, the first estimated increment value, and the second estimated increment value.

[0180] In this embodiment of the invention, the device further includes:

[0181] An interpolation processing module is used to determine if the bandwidth at the sampling time is abnormal for bandwidth data of each historical period. If the bandwidth at the sampling time is less than a third preset threshold, the bandwidth at the previous sampling time is greater than or equal to the third preset threshold, and the bandwidth at the next sampling time is greater than or equal to the third preset threshold, then the bandwidth at the sampling time is determined to be abnormal. The module then interpolates based on the bandwidth at the previous sampling time and the bandwidth at the next sampling time to determine the bandwidth at the sampling time.

[0182] A smoothing module is used to determine that the bandwidth at the sampling time is a sudden drop value if the bandwidth at the sampling time is greater than a fourth preset threshold, and the ratio of the difference between the bandwidth at the sampling time and the bandwidth at the previous sampling time to the bandwidth at the previous sampling time is greater than the fourth preset threshold, and then smooths the bandwidth at the sampling time.

[0183] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0184] This invention also provides an electronic device, comprising:

[0185] It includes a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, it implements the various processes of the above-described method embodiment for predicting the incremental service delivery network and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0186] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described method embodiment for predicting the incremental service volume of a content delivery network and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0187] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.

[0188] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, embodiments of the present invention can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of the present invention can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0189] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0190] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0191] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0192] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0193] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0194] The above provides a detailed description of the method, apparatus, device, and medium for predicting service increments in a content delivery network provided by the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for predicting incremental service usage in a content delivery network, characterized in that, The method includes: Acquire bandwidth data for multiple historical periods of the target service; wherein the bandwidth data for each historical period includes bandwidth data collected at multiple sampling times according to a preset sampling interval; Based on the bandwidth data from the multiple historical periods, the bandwidth data from the most recent historical period is selected as the reference bandwidth data; Determine the similarity between the reference bandwidth data and bandwidth data from other historical periods besides the reference bandwidth data; Based on the similarity, a preset number of bandwidth data for a target historical period are determined; The bandwidth data for the target historical period is weighted and averaged to determine the estimated bandwidth data for the next period; Based on the estimated bandwidth data for the next cycle, determine the target increment information for the next cycle; Determining the similarity between the reference bandwidth data and bandwidth data from other historical periods, excluding the reference bandwidth data, includes: Based on the dynamic time warping function, a first distance is calculated between the reference bandwidth data and the bandwidth data of other historical periods; the first distance is inversely correlated with the similarity. Based on the first distance, determine the similarity between the reference bandwidth data and bandwidth data from other historical periods besides the reference bandwidth data; The step of determining a preset number of bandwidth data for a target historical period based on the similarity includes: Determine the bandwidth data of the candidate historical period corresponding to the first distance that is less than or equal to the second preset threshold; The bandwidth data of the candidate historical periods are sorted in ascending order of the first distance; A preset number of candidate historical periods' bandwidth data are selected in the order described above, and these are used as the bandwidth data for the target historical period.

2. The method according to claim 1, characterized in that, The step of calculating the first distance between the reference bandwidth data and the bandwidth data of other historical periods based on the dynamic time warping function includes: Determine the sub-bandwidth data of the reference bandwidth data; Based on the dynamic time warping function, calculate the second distance between the sub-bandwidth data of the reference bandwidth data and the bandwidth data of other historical periods; If the second distance is less than or equal to the first preset threshold, then the reference bandwidth data is calculated, and the first distance between the reference bandwidth data and the bandwidth data of other historical periods corresponding to the second distance is calculated.

3. The method according to claim 1, characterized in that, The step of determining the target increment information for the next period based on the estimated bandwidth data for the next period includes: Based on the estimated bandwidth data for the next cycle, determine the first estimated increment value; Obtain the second estimated increment value; The target increment information for the next period is determined based on the first estimated increment value and the second estimated increment value.

4. The method according to claim 3, characterized in that, The step of determining the target increment information for the next period based on the first estimated increment value and the second estimated increment value includes: Obtain the estimated accuracy value and the business risk value; The target increment information for the next period is determined based on the estimated accuracy value, the business risk value, the first estimated increment value, and the second estimated increment value.

5. The method according to claim 1, characterized in that, Before selecting the bandwidth data from the most recent historical period as the reference bandwidth data based on the bandwidth data from the multiple historical periods, the method further includes: For bandwidth data of each historical period, if the bandwidth at the sampling time is less than the third preset threshold, the bandwidth at the previous sampling time is greater than or equal to the third preset threshold, and the bandwidth at the next sampling time is greater than or equal to the third preset threshold, then the bandwidth at the sampling time is determined to be abnormal. The bandwidth at the sampling time is determined by interpolation based on the bandwidth at the previous sampling time and the bandwidth at the next sampling time. If the bandwidth at the sampling time is greater than the fourth preset threshold, and the ratio of the difference between the bandwidth at the sampling time and the bandwidth at the previous sampling time to the bandwidth at the previous sampling time is greater than the fourth preset threshold, then the bandwidth at the sampling time is determined to be a sudden drop value, and the bandwidth at the sampling time is smoothed.

6. A service increment prediction device for a content delivery network, characterized in that, The device includes: A bandwidth data acquisition module is used to acquire bandwidth data for multiple historical periods of a target service; wherein the bandwidth data for each historical period includes bandwidth data collected at multiple sampling times according to a preset sampling interval. The reference bandwidth data determination module is used to select the bandwidth data of the most recent historical period as the reference bandwidth data based on the bandwidth data of the multiple historical periods. A similarity determination module is used to determine the similarity between the reference bandwidth data and bandwidth data from other historical periods other than the reference bandwidth data; The target historical period bandwidth data determination module is used to determine a preset number of target historical period bandwidth data based on the similarity. The estimated bandwidth data determination module is used to perform a weighted average of the bandwidth data of the target historical period to determine the estimated bandwidth data for the next period. The target increment information determination module is used to determine the target increment information for the next period based on the estimated bandwidth data for the next period. The similarity determination module includes: The first distance determination submodule is used to calculate a first distance between the reference bandwidth data and the bandwidth data of other historical periods based on a dynamic time warping function; the first distance is inversely correlated with the similarity. The similarity determination submodule is used to determine the similarity between the reference bandwidth data and bandwidth data from other historical periods other than the reference bandwidth data, based on the first distance. The first distance determination submodule includes: A sub-bandwidth data determination unit is used to determine the sub-bandwidth data of the reference bandwidth data; The second distance determination unit is used to calculate, according to the dynamic time warping function, the second distance between the sub-bandwidth data of the reference bandwidth data and the bandwidth data of other historical periods; The first distance determination unit is used to calculate the first distance between the reference bandwidth data and the bandwidth data of other historical periods corresponding to the second distance if the second distance is less than or equal to the first preset threshold.

7. An electronic device, characterized in that, include: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the steps of the service increment prediction method for a content delivery network as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the steps of the service increment prediction method for the content delivery network as described in any one of claims 1 to 5.

Images